Which Company Has the Second-Best AI Model in July?

The AI benchmarking race has never been tighter, and the second-place slot may actually be harder to call than first. Anthropic sits at exactly even odds to hold the second-best AI model position by end of July 2026, with the market split clean down the middle between Anthropic claiming the spot and one of thirteen other competitors taking it. That 50-50 split is a signal in itself: traders cannot agree on whether Claude 4 cements Anthropic’s standing or whether Google’s Gemini Ultra, xAI’s Grok, or DeepSeek’s latest model has already closed the gap.

The market asks which company holds the second-best AI model by July 31, 2026, with Anthropic priced at $0.50 (50% implied probability) and the combined field of alternatives priced at $0.50. Total volume stands at $4,166 with $18,033 in liquidity, making this a low-volume market where individual trades can shift the price meaningfully.

How the Anthropic Contract Works

This contract resolves YES if Anthropic holds the second-best AI model position by July 31, 2026, as determined by the resolution source. The specific benchmark or evaluator used for resolution matters enormously here, since different leaderboards (LMSYS Chatbot Arena, HELM, or independent third-party assessments) routinely produce different rankings. Resolution is binary: Anthropic either holds the second spot or it does not.

• YES at $0.50 (50%): Anthropic’s Claude models rank second globally among all AI providers by July 31, 2026.
• NO at $0.50 (50%): Any other company, including Google, OpenAI, xAI, DeepSeek, or Meta, holds the second-best AI model position on the resolution date.

A NO payout requires a rival to leapfrog Anthropic before the deadline. Google has the most credible path here: Gemini 2.5 Pro scored competitively on multiple benchmarks in early 2026, and Google’s compute advantages give it room to push another update before July 31. DeepSeek’s R2 or a successor model from xAI’s Grok line also represent plausible alternatives, particularly if Anthropic delays a major Claude update past the resolution window.

Market Signals: Dead Calm With a Volatile Day Behind It

Momentum on this contract is technically positive over the last hour (0.0% change) but the 24-hour data is unavailable, which limits clean composite analysis. The trend score of 12.50 is notably elevated, suggesting the market saw sharp activity recently. Price history confirms it: the contract swung down 16.5% and then recovered 14.5% on June 4, likely tied to a benchmark release or an AI announcement that briefly repriced Anthropic’s standing before traders faded the move. The contract returned to $0.50, which reads less like consensus and more like exhaustion.

Total volume of $4,166 with $18,033 in liquidity flags this as a thin market. The order book depth is real, but trading volume is low enough that a few informed traders with a strong view on June or July benchmark results could move this contract 10 to 15 cents without much resistance. That makes the current price less a conviction signal and more a placeholder.

• Anthropic sits at $0.50 (50%): the market has no edge and knows it.
• Trend score of 12.50: unusually high for a static price, pointing to recent churn and unresolved directional pressure.
• 1-hour change of 0.0%: activity has paused after yesterday’s swing, not stabilized.
• $4,166 total volume: thin enough that a single informed bet could reprice this contract before end of June.
• 24h volume equals total volume: this market essentially opened and traded almost entirely in the last 24 hours.

Lines Analysis: Anthropic Versus a Deep Field

Anthropic’s case for holding second place rests on Claude Sonnet 4 and Claude Opus 4, both of which earned strong scores on coding and reasoning benchmarks in mid-2026. Anthropic’s safety-focused development cycle tends to produce fewer rushed releases, and the company’s partnership with Amazon Web Services gives Claude broad enterprise deployment that reinforces its visibility in third-party evaluations. If the resolution source weights real-world usage and professional benchmarks, Anthropic’s deep integration across enterprise tools is a durable advantage through July.

The field threatens Anthropic on multiple fronts. Google’s Gemini 2.5 Pro has traded places with Claude on several leaderboards throughout 2026, and Google’s infrastructure scale means a targeted update before July 31 is operationally straightforward. DeepSeek’s rapid iteration cadence and xAI’s Grok series have both surprised benchmark watchers before. OpenAI occupies the market consensus for first place, which actually creates a narrow but real scenario where OpenAI slips and Anthropic inherits first, leaving second up for grabs entirely.

Signals to watch before July 31:

• Anthropic releases a Claude Opus 4.1 or equivalent model update that lands above current Claude scores on LMSYS or HELM, pushing YES toward $0.65.
• Google posts a Gemini 2.5 Ultra update or scores a high-profile deployment win that moves Gemini rankings above Claude, pulling YES toward $0.35.
• DeepSeek publishes a new model with comparable or superior reasoning scores to Claude Sonnet 4, adding a second credible threat to Anthropic’s position.
• xAI releases a major Grok update timed to a benchmark cycle, injecting volatility into this contract’s thin order book.
• The resolution source clarifies its methodology before July 31, which could sharply move this contract depending on whether the evaluator favors capability benchmarks or deployment-weighted assessments.

Total volume of $4,166 reflects genuine uncertainty, not deep research. The market is priced as a coin flip because traders lack a clear read on which benchmark the resolution will use and which model will lead that benchmark in late July. Both sides have real substance, and neither has a decisive edge as of June 5, 2026.

LINES VERDICT

TOO CLOSE TO CALL

Anthropic holds credible second-place standing today, but the resolution date is nearly two months out and Google’s Gemini has already traded places with Claude multiple times in 2026 alone.

What the market says: 50% implied probability with a trend score of 12.50 signals a market that saw sharp movement and snapped back to neutral, not one that has found conviction. Volatility will increase as July 31 approaches and benchmark releases clarify the standings.

The Competitive Landscape for AI Model Rankings

The second-place ranking in AI is genuinely contested in a way the top slot is not. OpenAI’s GPT-4o and o-series models have held the top position on most widely-cited benchmarks for most of 2025 and into 2026, giving the first-place question a cleaner answer. Second place is where Anthropic, Google, and increasingly DeepSeek and xAI are actively fighting over leaderboard position.

Benchmark methodology shapes outcomes as much as raw model capability. LMSYS Chatbot Arena uses human preference voting, which tends to favor conversational fluency and can favor Claude’s style. Academic benchmarks like MMLU and HumanEval favor raw reasoning and coding accuracy, where Google’s Gemini and DeepSeek have been competitive. If the resolution source uses a composite or a single methodology, the specific choice will heavily influence which company holds second by July 31.

Events that move this market before the deadline include any major model release from Anthropic, Google, DeepSeek, or xAI; a shift in LMSYS leaderboard rankings; a published third-party evaluation that names a clear second-place model; or a resolution source announcement that clarifies the evaluation framework.